By Ben Ellison
Circles of Uncertainty
|Part 2: So-called “overzooming” on our plotter or PC may make the details easier for us to see, but doesn’t improve the accuracy a bit.|
Old-fashioned data is but one problem with new-age charting. Calder goes into fascinating detail about how 19th-century scientists struggled to mathematically define the earth for the sake of accurate surveying. You may recall that our planet is not perfectly round, but did you realize that it defies mathematical modeling? The upshot is that there are a wide variety of “Horizontal Datums” (earth models) that work fine relative to nearby features and even celestial observations but don’t blend perfectly into one global system. While a slightly more vague but global datum (WGS84) is being adopted in the age of GPS, not all paper charts have been switched over. In short, it’s wrong to transfer a GPS (WGS84) position to a chart with an older datum, or vice versa—as much as 330 feet wrong with some NAD27 U.S. charts only five years old, 3,000 feet wrong with some foreign datums.
While this problem, if understood, is easily solved by switching a GPS to the paper chart’s datum version, there are other annoyances. Notice in that screen shot (right) how a five-foot sounding on the original chart somehow became a harder-to-read 4'9" during vectorization. Similarly, while the “pre-1900’s” source information I found on my local paper charts (and their raster equivalents) is supposedly embedded into most vector charts, I don’t know of a recreational plotter or charting program that shows it. But what really worries Calder, and many marine safety experts, is how all the “little” issues of accuracy and translation get multiplied when electronic charts are magnified, even smoothed out, on our screens. Old data aside, Calder calculates that the 1:40,000 raster chart (above right) is only accurate to about 1⁄16 inch on paper, 66 yards in real life. Plus bear in mind that the cartographer may have left out redundant ledges or widened channels for legibility on the paper. So-called “overzooming” on our plotter or PC may make the details easier for us to see, but doesn’t improve the accuracy a bit. And while the act of putting the underlying paper chart under a microscope is obvious with the raster format, the vector version—the exact same data except with possible added tracing errors—looks great magnified. The temptation to go where perhaps you shouldn’t, or faster than prudent, is great.
You’ve been warned. And, by the way, it’s predicted that the sub captain will loose his stripes due to overconfident speed in an area that was known to be poorly surveyed. I highly recommend reading Calder’s book, chock-a-block with many more pointers than I’ve been able to cover here. Cultivate a sense of uncertainty. On most plotters and programs, you can even plot a literal circle of uncertainty using the vessel range function. The idea is not to quit exploring, but to minimize surprises. Note that Calder’s latest boat is equipped with a 15-inch cockpit monitor and some very snazzy electronic charting. I picture him using it to gunkhole—even bump around a bit—with great gusto, but also with care. It’s a smart way to cruise.
This article originally appeared in the April 2005 issue of Power & Motoryacht magazine.